Releases

https://libvirt.org/news.html

v3.10.0 (unreleased)
    qemu: Add vmcoreinfo feature
        Starting with QEMU 2.11, the guest can save kernel debug details when this feature is enabled and the kernel supports it.
3.9.0 (2017-11-02) - Debian 10 buster / sid
v3.8.0 (2017-10-04)
v3.7.0 (2017-09-04)
    bhyve: Support autoport for VNC ports
    qemu: Added support for setting heads of virtio GPU
v3.6.0 (2017-08-02) - Ubuntu 17.10 artful / 18.04 bionic
    Handle hotplug change on VLAN configuration using OVS
    fix some minor bugs when using 'host-model' CPU.
v3.0.0 (2017-01-17) - Debian 9 stretch & 8 jessie-backports
v2.5.0 (2016-12-04) - Ubuntu 17.04 zesty

install

https://help.ubuntu.com/community/KVM/Installation

kvm-ok
apt-get install qemu-kvm libvirt-bin ubuntu-vm-builder bridge-utils
systemctl status libvirtd

Config

cd /etc/libvirt/qemu && virsh define foo.xml

https://wiki.libvirt.org/page/FAQ#Where_are_VM_config_files_stored.3F_How_do_I_edit_a_VM.27s_XML_config.3F

Debug

https://fedoraproject.org/wiki/How_to_debug_Virtualization_problems

virt-host-validate
virsh capabilities

http://events.linuxfoundation.org/sites/events/files/slides/Debugging-libvirt-QEMU-in-OpenStack-2015-CloudOpen-Eu.pdf

/etc/libvirt/libvirtd.conf
    log_filters="1:qemu 1:libvirt 3:security 3:event 3:util 3:file" log_outputs="1:file:/var/log/libvirt/libvirtd.log"
systemctl restart libvirtd

https://libvirt.org/logging.html

export LIBVIRT_DEBUG=1
export LIBVIRT_LOG_FILTERS="1:qemu 1:libvirt"
export LIBVIRT_LOG_OUTPUTS="1:journald 1:file:/tmp/virsh.log"

LIBVIRT_LOG_FILTERS:
    x:name  (log message only)
    x:+name (log message + stack trace)
    1: DEBUG
    2: INFO
    3: WARNING
    4: ERROR

LIBVIRT_LOG_OUTPUTS
    stderr output goes to stderr
    syslog:name use syslog for the output and use the given name as the ident
    file:file_path output to a file, with the given filepath
    journald output goes to systemd journal

gdb: https://access.redhat.com/blogs/766093/posts/2690881

virsh

virsh list --all
virsh destroy ... # force off
export EDITOR=vim # for `virsh edit`
virsh dumpxml GuestID > guest.xml
virsh define  guest.xml
virsh domxml-to-native qemu-argv demo.xml > demo.sh
virsh domxml-from-native qemu-argv demo.args > demo.xml

snapshot

virsh snapshot-create-as --domain debian --name 'snapshot1'
virsh snapshot-list debian
qemu-img snapshot -l debian.qcow2 # when VM off

Pool

virsh pool-edit default     # debian 9 default: /var/lib/libvirt/images/

Driver

https://fedorapeople.org/groups/virt/virtio-win/direct-downloads/stable-virtio/virtio-win.iso
https://github.com/torvalds/linux/blob/master/drivers/virtio/Kconfig
VIRTIO_BALLOON supports increasing and decreasing the amount of memory within a KVM guest.
VIRTIO_MMIO support for memory mapped virtio platform device driver.

OVS

http://docs.openvswitch.org/en/latest/howto/libvirt/ - virt-manager / create bridge :

MacVTap

https://hicu.be/bridge-vs-macvlan

  • Macvlan Bridge: VM-VM has bridge
  • Macvlan private: VM-VM no communication
  • Macvlan VEPA: VM-VM communicate via VEPA capable switch

https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Virtualization_Administration_Guide/sect-attch-nic-physdev.html
passthrough attaches a virtual function of a SRIOV capable NIC directly to a VM without losing the migration capability.

https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Virtualization_Host_Configuration_and_Guest_Installation_Guide/App_Macvtap.html
when a guest virtual machine is configured to use a type=’direct’ network interface such as macvtap, …, the guest cannot communicate with its own host.

My workaround: deassign macvtap host NIC’s IP, and communicate with host’s 2nd NIC.

Linux bridge

try if no dhcp:

sysctl -w net.ipv4.conf.br0.bc_forwarding=1
sysctl -w net.bridge.bridge-nf-call-iptables=0

VLAN

Macvlan Bridge

ip link set dev enp4s0 promisc on

OVS - VLAN tagging transparent

https://libvirt.org/formatdomain.html#setting-vlan-tag-on-supported-network-types-only

<devices>
<interface type='bridge'>
    <vlan>
        <tag id='42'/>
    </vlan>
    <source bridge='ovsbr0'/>

Shutdown timeout

vi /etc/init.d/libvirt-guests
    ON_SHUTDOWN=shutdown
    SHUTDOWN_TIMEOUT=10

systemctl enable libvirt-guests.service

Nested

booting with kvm-intel.nested=1 argument on the kernel command line, or:
https://www.linux-kvm.org/page/Nested_Guests#How_to_run

virt-manager:enable Copy host CPU configuration checkbox

virtiofs

https://libvirt.org/kbase/virtiofs.html

vfio

https://www.kernel.org/doc/Documentation/vfio.txt

echo '0000:42:00.1' | sudo tee /sys/bus/pci/devices/0000:42:00.1/driver/unbind
echo 8086 1d02 | sudo tee /sys/bus/pci/drivers/vfio-pci/new_id

Raw disk mapping

Support partion

<disk type='block' device='disk'>
    <source dev='/dev/sdbX'/>    
    <target dev='vdb' bus='virtio'/>
</disk>

    <source dev='/dev/disk/by-uuid/abcd1234'/>

GRUB

lspci -knn | grep VGA -A 5
cat /etc/default/grub | grep GRUB_CMDLINE_LINUX_DEFAULT
    intel_iommu=on kvm-intel.nested=1 modprobe.blacklist=megaraid_sas,ahci,radeon,nouveau,nvidiafb,snd_hda_intel,...
update-grub

UEFI

apt install -y ovmf
systemctl restart libvirtd
# select UEFI while creating VM

Troubleshooting

error while adding virt-manager NFS pool:
    libvirtd[...]: cannot open volume '...': Transport endpoint is not connected
=> chmod +r